5 Steps to Improving Plant Reliability with Strategic Maintenance

Design a maintenance strategy that maximizes the availability of vital equipment, says Chris Allmond, Head of Engineering Consultancy Services at RS Integrated Supply

While developments in technology have long helped cut the cost of predictive maintenance, the pace of change has accelerated significantly over the last couple of decades. Now you can, for example, install vibration sensors to detect even a slight deviation from the usual parameters and automatically generate a text message or alert on your computerized maintenance management system.

People introduce technology because they like the bells and whistles, but while having sensors all over your factory telling you what is happening is all well and good, to make the most of this information you need to be able to use it proactively. You need a strategy and the correct processes as well as stakeholder buy in.

Here Chris Allmond, Head of Engineering Consultancy Services for RS Integrated Supply, explains how to develop this strategy and processes to sustainably improve reliability in a manufacturing environment.

When I go into a business and people tell me that their plant is unreliable, I always ask about compliance with their preventative maintenance program. They’re often chasing maximum production because of demanding performance targets at the expense of inspection, but the analogy I use is if I don’t get my car serviced, the warranty becomes null and void – and the same applies in industrial settings. In these scenarios, it’s the validity and value of maintenance inspections that needs addressing first. Go back and start doing what works. Ensure the basics happen: clean, tighten, lubricate, and maintain.

To understand your process productivity limitations, the recommended approach is to conduct a value stream mapping exercise, looking at the performance design criteria of the equipment. Map out every aspect of the process and you’ll get a holistic view of where there are constraints and limiting productivity factors. Realistically, this is what determines total available productivity.

You can then extrapolate how available equipment needs to be to fulfil the production demand. This is pretty telling, as equipment cannot be technically available 100% of the time without running the machinery into the ground.

Engineers can approach reliability centered maintenance in a similar way, starting with the primary and secondary function of machinery. To give a simplified example, a pump is installed to meet a set of specifications: to pump to a certain pressure or flow rate. That is its primary function. Its secondary functions are to contain whatever it is pumping and protect people from moving parts.

Look at functional failure and you’ll see the consequences of failure and also failure modes. Understanding failure modes allows you to introduce preventive or predictive criteria that will flag up early warning signs. You can then act quickly before the failure occurs, avoiding unnecessary and costly maintenance while optimizing condition and availability.

Thanks to Artificial Intelligence (AI), we now have smart systems that are hugely sophisticated and will monitor far more parameters than any human could. These smart systems are self-learning so will look for any deviation from a steady state, which can indicate the root cause of failures and ultimately support in protecting the assets.

Many businesses are moving in this direction. How I look at it is that installing sensors in the right places based on failure modes and the design criteria can eliminate some of the downtime needed for maintenance, giving you back that time as extra capacity on the production line. Trying to gain the same amount of additional capacity by buying new equipment but sticking with the same invasive maintenance strategies is more expensive.

There will always be situations in which you need to make an exception to your usual maintenance procedures. If an urgent order comes through, for instance, it may make commercial sense to keep a piece of vital machinery operating even if it was due to go offline for maintenance. In such scenarios, ensure all relevant stakeholders are involved in decision making as your engineers should have the data to know if deferring maintenance could result in a serious failure and, ultimately, more expense and disruption.

In some cases, letting a component fail may be the most appropriate tactic. Say you have multiple lights in an office block. You are not going to use sensors to detect when one of them is going to stop working as the cost and consequences of failure are not severe enough. Only when a number of them are out do you need to undertake remedial action.

That said, one area where there are no exceptions is safety. When you’re looking at equipment to define criticality on any site for any process, safety of people and the environment is key. With some core statutory inspections, you may be able to defer for up to six months, but to do so you must conduct a risk-based inspection to demonstrate that you can postpone without any negative impact. Safety is always the priority.

Imagine all of the machinery and equipment positioned along a sliding scale, with different tactics to mitigate failure according to their criticality. To ensure your overall maintenance strategy remains effective, you will need to amend these tactics over time to counteract degradation and new failure modes as equipment ages.

For the most vital equipment, for example, you might have a standby available. As soon as the standby is put into use, its criticality increases hugely because now you no longer have a spare. If the mean time between failures on a vital piece of equipment shoots up, this indicates that current maintenance tactics are not working. You need to investigate the underlying cause of the problem and find a solution.

With so many moving parts in engineering, you need this holistic approach that allows you to monitor everything that is going on across your operations – especially as organizations increasingly press their maintenance engineering teams to deliver more results from ageing equipment while optimizing resources.